Right-wing populist parties
and the rise of hate crimes

DVPW 2024, Göttingen

Heike Klüver

Humboldt-Universität zu Berlin

Jasper Jansen

Massachusetts Institute of Technology

Tim Wappenhans

Humboldt-Universität zu Berlin

September 25, 2024

Motivation

Hate crimes on the rise

news_01
The Guardian
news_02
Associated Press
news_03
Deutsche Welle

Preview





Does local right-wing party representation induce hate crimes?

Preview


Does local right-wing party representation induce hate crimes?

Design

  • DiD
  • German municipalities

Results

  • no effect
  • normalization
    • national/state level > local level

Theory

Argument

Social identity theory

  • in-group
    • comply with norms
  • out-group
    • perceived as threat

Populist radical right parties

  • nativism
    • the nation
  • populism
    • the people

Creation of distinct in-group, under attack by threatening out-group ⚡

Case

German municipalities

  • lowest administrative level
  • most direct point of contact
  • potential for normalization of hate

Data

Hate crimes in Germany

Challenge

Changing municipality borders

Measurement

Units

  • 6,390 (synthetic) municipalities, 2010-2020
  • 68,475 municipality*years


Outcome

  • right-wing hate crimes per 100,000

Treatment

  • AfD local election results

Hate crimes

Hate crimes

Hate crimes

Hate crimes

Identification

Identification

Estimand

  • \(\Delta\) right-wing hate crimes AfD present vs. not


Design

  • Staggered DiD

Treatment

  • AfD presence in council

AfD municipality presence

Results

Aggregated ATT

Exploring heterogeneity

Results may be different for

  • treatment cohorts

  • ideological motivation

  • treatment intensities

  • violent crimes

Treatment cohorts

Ideological motivation


Treatment intensity


Violent crimes

Conclusion

Takeaway 🥡


  • local PRR presence does not increase hate crimes

    • ideological motivation ❌
    • treatment intenstiy ❌
    • violent crimes ❌

  • normalization: national/state level > local level

Next steps

What do do next

  • collecting missing data for remaining states (may be a sample effect)
  • addressing multilevel nature of treatment: local, state, national level
  • further ideas welcome




Hit us up 📬

heike.kluever@hu-berlin.de

jasperj@mit.edu

tim.wappenhans@hu-berlin.de

Appendix

Event study